Conditional probability analysis of multidrug resistance in Gram-negative bacilli isolated from tertiary medical institutions in South Korea during 1999-2009.

Multidrug resistance of Gram-negative bacilli is a major problem globally. However, little is known about the combined probability of resistance to various antibiotics. In this study, minimum inhibitory concentrations of widely used antibiotics were determined using clinical isolates of Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Acinetobacter baumannii, randomly chosen from strain collections created during 1999-2009 in tertiary medical institutions in Seoul, South Korea. To analyze combined efficacy of antibiotics against a subgroup of isolates, conditional probabilities were determined based on arbitrary, non-independent patterns of antimicrobial susceptibility and resistance. Multidrug resistance, defined as resistance to three or more classes of antibiotics, was observed in the following order: A. baumannii (96%), P. aeruginosa (65%), E. coli (52%), and K. pneumoniae (7%). A. baumannii strains resistant to gentamicin were found to be resistant to a number of antibiotics, except for colistin and polymyxin B. Resistance to gentamicin following exposure to this antibiotic was highly likely to lead to multidrug resistance in all four microbes. This study shows a causal relationship between gentamicin resistance and the prevalence of multidrug resistance in clinical isolates of Gramnegative bacilli in South Korea during 1999-2009 and suggests the importance of prudent use of gentamicin in hospitals.